Heart Rate Variability and Firstbeat Method for Detecting Sleep Stages in Healthy Young Adults: Feasibility Study
Bibliographic record
Abstract
BACKGROUND: Polysomnography (PSG) is considered the only reliable way to distinguish between different sleep stages. Wearable devices provide objective markers of sleep; however, these devices often rely only on accelerometer data, which do not enable reliable sleep stage detection. The alteration between sleep stages correlates with changes in physiological measures such as heart rate variability (HRV). Utilizing HRV measures may thus increase accuracy in wearable algorithms. OBJECTIVE: We examined the validity of the Firstbeat sleep analysis method, which is based on HRV and accelerometer measurements. The Firstbeat method was compared against PSG in a sample of healthy adults. Our aim was to evaluate how well Firstbeat distinguishes sleep stages, and which stages are most accurately detected with this method. METHODS: Twenty healthy adults (mean age 24.5 years, SD 3.5, range 20-37 years; 50% women) wore a Firstbeat Bodyguard 2 measurement device and a Geneactiv actigraph, along with taking ambulatory SomnoMedics PSG measurements for two consecutive nights, resulting in 40 nights of sleep comparisons. We compared the measures of sleep onset, wake, combined stage 1 and stage 2 (light sleep), stage 3 (slow wave sleep), and rapid eye movement (REM) sleep between Firstbeat and PSG. We calculated the sensitivity, specificity, and accuracy from the 30-second epoch-by-epoch data. RESULTS: In detecting wake, Firstbeat yielded good specificity (0.77), and excellent sensitivity (0.95) and accuracy (0.93) against PSG. Light sleep was detected with 0.69 specificity, 0.67 sensitivity, and 0.69 accuracy. Slow wave sleep was detected with 0.91 specificity, 0.72 sensitivity, and 0.87 accuracy. REM sleep was detected with 0.92 specificity, 0.60 sensitivity, and 0.84 accuracy. There were two measures that differed significantly between Firstbeat and PSG: Firstbeat underestimated REM sleep (mean 18 minutes, P=.03) and overestimated wake time (mean 14 minutes, P<.001). CONCLUSIONS: This study supports utilizing HRV alongside an accelerometer as a means for distinguishing sleep from wake and for identifying sleep stages. The Firstbeat method was able to detect light sleep and slow wave sleep with no statistically significant difference to PSG. Firstbeat underestimated REM sleep and overestimated wake time. This study suggests that Firstbeat is a feasible method with sufficient validity to measure nocturnal sleep stage variation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".